Previous studies have indicated that oral epithelial dysplasia (OED) represents a condition with an elevated risk of transformation to oral cancer. However, while the epidemiology of oral cancer has received considerable attention, and while a limited number of studies have investigated risk factors for oral leukoplakia in general, little is known regarding the analytic epidemiology of OED. The proposed pilot study represents an important first step in exploring and quantifying the associations between OED and its suspected risk factors. The primary aim of the proposed study is to measure the independent association between oral epithelial dysplasia and the use of smoking tobacco and alcoholic beverages. Secondarily, and on an exploratory basis, the investigation will also (a) examine for the presence of a dose-response relationship between OED and both tobacco and alcohol use, (b) test for a synergistic relationship between alcohol and smoking tobacco use as regards OED, (c) evaluate smokeless tobacco use, mouthwash use and denture status as independent risk factors for OED, and (d) evaluate various dietary habits as risk factors for OED. In this case-control study the dependent variable of interest is histologically confirmed epithelial dysplasia of the oral cavity, and the primary independent variables of interest are the current and past use of alcoholic beverages and smokinq tobacco. Cases will be identified by reviewing biopsy reports from a major oral pathology laboratory serving oral surgeons and general dentists practicing in either Connecticut or Massachusetts. Controls, identified through practices giving rise to the identified cases, will be pair-matched (1:1) to cases on age (equal to or more than 5 years) and gender. Exposure status as regards the potential risk factors discussed above will be ascertained through a written food frequency questionnaire and a telephone interview. Estimates of the relative risk of tobacco use and alcohol consumption as regards OED will be based upon adjusted odds ratios from a conditional logistic regression model in which relevant covariates are included. On an exploratory basis, conditional logistic regression will also be used to test for (a) a dose-response relationship between OED and both tobacco use and alcohol consumption and (b) a synergistic effect between smoking tobacco use and alcohol consumption as regards OED. Finally, conditional logistic regression will be used to evaluate smokeless tobacco use, mouthwash use, denture wearing, and various dietary habits as independent risk factors for oral epithelial dysplasia. In keeping with NCI small grant criteria, this pilot study will both (a) evaluate the feasibility of using dental offices as a source of cases and controls for studies of oral cancer and precancer, and (b) provide initial estimates for associations between OED and its suspected risk factors.